Emergence and variability of influence exerted by weather and climatic factors on life expectancy in the Russian Federation taking into account differentiation of RF regions as per socioeconomic and sanitary-epidemiologic determinants

The present research focuses on estimating influence exerted by weather and climatic factors on life expectancy (LE) in the Russian Federation taking into account socioeconomic and sanitary-epidemiologic determinants. To estimate influence exerted by this factor on LE, a mathematic model was applied; the model was based on neuron networks and allowed taking into account emergence and variability of influence exerted on changes in LE by a set of heterogeneous factors including weather and climatic ones. It was established that over 2010-2018 climate changed in most RF regions as there was a growth in average monthly temperatures (temperature deviated from its long-term average monthly values by +1.2 ºС in July, and by +1,5 ºС in January),and changes in precipitations (deviations amounted to -1.9% in July and +13.0 % in January). It was established that «average monthly temperature in July» exerted the greatest direct influence on LE; thus, if this parameter grows by 1 %, it results in additional 1.7 days of LE. «Average precipitations quantity in January» turned out to be the most significant factor leading to a decrease in LE; a 1 % growth in this parameter resulted in LE decrease by 0.12 days. It was shown that mathematical expectancy of LE loss variability in RF regions obtained basing on 85 scenarios of weather and climatic conditions ranged from -4.2 days to 348.7 days. Overall in the RF climate-associated losses in LE taken as weighted average as per population number amounted to 191.7 days. It was established that climate-associated losses in LE were authentically lower in North Caucasian regions than in regions located in temperate zone with Atlantic-continental and continental climate (by 1.6 and 1.8 times accordingly). We also comparatively analyzed losses in LE due to influence exerted by climate in RF regions distributed into different groups (clusters) as per socioeconomic parameters; the analysis revealed authentic differences between the second and the fourth cluster (p=0.01), and between the third and the fourth ones (p=0.006). We didn’t reveal any authentic differences in climate-associated losses in LE among clusters as per sanitary-epidemiologic parameters.